Introduction
The sensory system is a part of the nervous system that processes sensory information, which includes receptors, neural pathways, and the sensory center of the cerebral cortex. In the sensory neural network of humankind, the sensory receptors convert environmental information into potential changes and encode such potential changes into spike trains with neural spike coding in the cell body. Subsequently, interneurons convey the spike trains from the receptors to the cerebral cortex of the brain, where the information is decoded into sensory perceptions for further processing.[1-6] Such structure forms the basis of sensing, pre-processing and encoding capabilities of the human sensory system. This outperforms digital computers in dealing with a large number of complex tasks, such as perception and real-time sensory data processing. Therefore, it is of great significance to draw inspiration from the human sensory system and develop artificial sensory hardware that can efficiently realize the perception and encoding capabilities, which will provide a promising approach towards e-skin, neurorobotics and human-machine interaction technologies.
Traditional complementary metal oxide semiconductor (CMOS) technology utilizes complex auxiliary circuits and bulky capacitors to emulate sensors and bio-dynamics, which takes up a large area and high computational cost. [6-10] In recent years, many researchers have devoted efforts to emulating the synaptic dynamics or neuronal behaviors using emerging devices such as non-volatile memristors, [11-13] volatile memristors [14, 15] and synaptic transistors.[16, 17] Meanwhile, there were recent reports on emulation of biological sensing functions by integrating functional sensors with synaptic and neuron components. [18-28] One way is to use sensors combined with artificial synapse devices, [19-27] and recently, volatile memristors have emerged as excellent candidates for the construction of artificial sensory neurons due to their simple two-terminal structure and dynamic threshold switching (TS) characteristics. [18, 23, 28, 29, 32] However, aiming at sensory neuron, these prior studies only focus on single-mode sensory perception, but rarely achieves the integration of multiple sensory inputs, [28-32]which is distant from the efficient processing of cross-sensory information in biology. Therefore, it is of great significance to construct an artificial sensory system that not only senses and converts the physical information in real time, but also directly fuse and integrate multisensory inputs in a hardware-efficient manner.
Here, we report an artificial multisensory neuron consisting of a piezoresistive sensor and a VO2 based volatile memristor connected in series. Such artificial sensory neurons can be used to sense different pressure inputs and convert them into spike trains as a result of the voltage dividing effect between the piezoresistive sensor and VO2 memristor. Besides, Tthe spiking neuron is also capable of sensing temperature, by taking advantage of the intrinsic thermal sensitivity of metal-insulator transition in VO2. The spiking neuron is utilized to recognize Braille characters using multiple piezoresistive sensors. Notably, the traditionally separate haptic and temperature signals can now be fused physically in the VO2 based sensory neuron when synchronizing the two sensory cues, which is able to recognize multimodal haptic/temperature patterns. Such multisensory neurons could provide a promising approach towards e-skin, neuro-robotics and human-machine interaction technologies.